A recent tritium inventory imbalance at Mound required an investigation into its causes. Much tritium was found as HTO in unsuspected zeolite traps of a T-purification system. Isotopic exchange from ammonia was postulated as a mechanism for entry of T into the zeolitic water. Gases from a T-processing system which had experienced air in-leakage were shown, by trapping of condensibles, to contain substantial H-isotopic waters and ammonias. Further evidence for tritiated ammonia was inferred from changes in pressure and T purity in otherwise unperturbed tanks of N2 and (H/D/T)2. From two such tanks which held N2 and T2 at equilibrium, ammonia was trapped and decomposed; a preliminary equilibrium constant for N2 + 3T2 ⇔ 2NT3 was determined. Controlled experiments by laser Raman spectrometry are in progress to investigate N2/T2 equilibria and kinetics. Results for gas mixtures in the 60–130 kPa (500–1000 torr) range (per reactant) suggest that the forward rate and the equlibrium attained are α [T2]2 . G-values for NT3 production were ≈1–2 molecules atm-1 (T2) (100eV)-1. Self-decomposition of NT3 proceded in an exponential decay with a G = 15–30. A lower value was observed at pressures where β--absorption in the gas was poor.